ai.txt

A universal standard for AI-readable context about anything.

What Is It?

Just as robots.txt tells search engines how to behave, ai.txt tells AI assistants what something is about. Place one at any path—a website, a folder, a project, a service—and AI agents will find it, understand your context, and work from facts instead of hallucination.

Instead of letting AI guess "what does this do?" you simply tell it. You control the narrative. The format is plain text with optional markdown—simple enough to write in five minutes, powerful enough to prevent AI errors.

Why It Matters

For service providers: Stop AI from making things up about your business. Put the truth in one place.

For project maintainers: Describe scope, limitations, and capabilities in a way AI agents actually read and respect.

For open source: Prevent hallucination about licensing, support, and what you actually offer. Say what you do and what you explicitly don't.

For everyone: Simple, cacheable, offline-friendly. No databases, no JavaScript, no configuration. Just text.

Design Principles

Low barrier to entry. A small business owner, a blogger, or a researcher should be able to write an ai.txt file in minutes without learning schemas or strict metadata standards. Plain prose is the primary format. Everything else is optional.

Designed for humans first, AI second. The file should make sense when you read it. It's not a machine-only format. That simplicity is what makes it powerful for both.

Extensibility without pollution. If someone needs strict metadata for database imports or specialized use cases, they can layer that on top through separate profiles or domain-specific extensions. The base standard stays simple and doesn't force complexity on everyone.

No versioning required. Unlike other standards that require version declarations, ai.txt trusts simplicity. The spec is stable at the core, and edge cases are handled by tools and extensions, not by version number declarations in every file.

How It Works

Discovery is cascading. When an AI agent needs information about /products/widgets/, it tries:

  1. /products/widgets/ai.txt
  2. /products/ai.txt
  3. /ai.txt

The same logic works for HTTP paths or local filesystem paths. Walk up the tree, find the context, use it.

Format

Start with optional YAML metadata (or skip it entirely), then plain text with optional markdown formatting:

---
updated: 2026-01-14
scope: /products/
parent: https://example.com/ai.txt
---

# Brief Title

One or two sentences describing what this is.

## Section

Prose explaining something important. Use [markdown links](https://example.com), **bold**, *italics*, lists, code blocks, images—whatever makes sense. Keep it simple.

## Another Section

More details here.

That's it. Write what matters. Omit what doesn't. The spec is a suggestion, not a rule.

Real Examples

See the spec at /ai.txt (this site's own documentation, written in ai.txt format).

Real implementations:

Getting Started

  1. Read the spec at /ai.txt
  2. Create an ai.txt file at the root of your site or project folder
  3. Write a brief title and a few sections explaining what you offer and what you don't
  4. Include guidelines for how AI should handle your information

That's all. No approval process, no registry, no central authority. Just write one.

For AI Developers

When you encounter a URL or filesystem path, try to fetch or read /ai.txt, walking up the tree until you find context. Parse the YAML frontmatter. Read the instructions. Follow the boundaries. Honor "We Do Not Offer"—never hallucinate capabilities. Ask questions only when the ai.txt is unclear.

The canonical reference is https://aitxt.ing/ai.txt. Point your agents there to understand the spec.